When two influenza A viruses (IAVs) co-infect the same cell, exchange of gene segments through reassortment gives rise to up to 256 different viral variants. This process has the potential to bring together viral genes that have evolved in distinct host species, which in turn can facilitate zoonotic transfer. In a host infected with a single virus population, reassortment can furthermore couple beneficial mutations or purge deleterious ones, thereby accelerating viral evolution. Using a unique experimental system that allows direct quantification of reassortment, we have shown that this process of genetic exchange is highly efficient in cell culture and in a guinea pig host. Indeed, the prevalence of reassortant viruses shed from co-infected guinea pigs (up to 75%) suggests that reassortment is a routine feature of IAV infections in nature. IAVs circulate in a wide range of avian and mammalian species, however, and it remains unclear whether our findings in guinea pigs would extend to these natural hosts. We therefore propose to evaluate the efficiency of reassortment in pigs, mallards, quail and chickens. Ferrets and guinea pigs will also be included as models of human IAV infection. By comparing results among species we will not only gain important insight into the incidence of reassortment in natural hosts, but also test the hypothesis that the efficiency of reassortment varies among host species. To investigate the mechanisms determining IAV reassortment efficiency in vivo, we will use immunohistochemistry to track the spread of co-infecting viruses over a time course after inoculation. We hypothesize that reassortment is favored by focal spread within target tissues, creating local areas where cells are infected at high multiplicity. Finally, since our work in cell culture has revealed that viral genomes lacking one or more segments markedly enhance the frequency of reassortment, we will test whether the prevalence of incomplete viral genomes varies with host species. Our preliminary data strongly suggest that missing segments are lost, not during virion assembly, but rather after genomes are delivered to a new target cell. The likelihood of segments reaching the site of replication furthermore appears to depend on the species from which the target cell is derived. Thus, we propose that reassortment levels may be host dependent in part due to differing frequencies of incomplete viral genomes and will test this novel concept through experimentation both in cell culture and in animal models. Overall, the research proposed will deepen mechanistic understanding of IAV reassortment, a critical process in the diversification, evolution and emergence of novel influenza A viruses.